openpyxlで数万〜数十万行のエクセルを作る際の時間を調べてみた（Lambda）

sam init \
    --runtime python3.11 \
    --name lambda-openpyxl-sample \
    --app-template hello-world \
    --no-tracing \
    --no-application-insights \
    --structured-logging \
    --package-type Zip

openpyxl

AWSTemplateFormatVersion: '2010-09-09'
Transform: AWS::Serverless-2016-10-31
Description: lambda-openpyxl-sample

Resources:
  HelloWorldFunction:
    Type: AWS::Serverless::Function
    Properties:
      FunctionName: lambda-openpyxl-sample-function
      CodeUri: hello_world/
      Handler: app.lambda_handler
      Runtime: python3.11
      MemorySize: 8192
      Timeout: 900
      Architectures:
      - x86_64

  HelloWorldFunctionLogGroup:
    Type: AWS::Logs::LogGroup
    Properties:
      LogGroupName: !Sub /aws/lambda/${HelloWorldFunction}

import openpyxl
import time

def lambda_handler(event, context):
    wb = openpyxl.Workbook()
    ws = wb.active

    # 50万行 x 10列のデータを書き込む
    begin_write = time.perf_counter()
    for i in range(500_000):
        ws.append([f"Row {i}"] + [f"Column {j}" for j in range(1, 10)])
    end_write = time.perf_counter()
    print(f"Write time: {(end_write - begin_write)} s")

    begin_save = time.perf_counter()
    wb.save("/tmp/hello_world.xlsx")
    end_save = time.perf_counter()
    print(f"Save time: {end_save - begin_save} s")

sam build --use-container
sam deploy

aws lambda invoke \
    --function-name lambda-openpyxl-sample-function \
    output.txt

import openpyxl
import time

def lambda_handler(event, context):
    wb = openpyxl.Workbook(write_only=True)
    ws = wb.create_sheet()

    begin_write = time.perf_counter()
    for i in range(500_000):
        ws.append([f"Row {i}"] + [f"Column {j}" for j in range(1, 10)])
    end_write = time.perf_counter()
    print(f"Write time: {(end_write - begin_write)} s")

    begin_save = time.perf_counter()
    wb.save("/tmp/hello_world.xlsx")
    end_save = time.perf_counter()
    print(f"Save time: {end_save - begin_save} s")